Enhanced multi-omic viral profiling from microbial community sequencing with BAQLaVa

Area of Science:

• Microbiology
• Virology
• Bioinformatics

Background:

• Viruses are essential but often undetected components of microbial ecosystems.
• Current technologies face limitations in experimental and analytical viral detection.
• Understanding viral communities is crucial for human health and disease research.

Purpose of the Study:

• To develop a novel algorithm, BAQLaVa, for high-resolution viral profiling.
• To improve the detection and analysis of viral species in complex biological samples.
• To investigate viral community dynamics in inflammatory bowel diseases (IBD).

Main Methods:

• Developed BAQLaVa algorithm for viral genome bin (VGB) profiling using reference-based alignment.
• Utilized nucleotide markers and proteome sets for enhanced viral detection.
• Applied BAQLaVa to metagenome (MGX) and metatranscriptome (MTX) samples from the HMP2 IBDMDB cohort.

Main Results:

• BAQLaVa achieved >90% species-level recall and precision, outperforming existing methods.
• Identified reduced virome diversity correlating with bacterial diversity during IBD inflammation.
• Discovered enrichment of specific viruses in IBD, associated with anti-defense systems and lysogeny regulation.
• Inferred phage-host relationships using scalable co-occurrence and covariation signals.

Conclusions:

• BAQLaVa offers a scalable framework for sensitive and specific viral profiling from MGX and MTX data.
• The algorithm facilitates virome epidemiology and systematic analysis of virus-host interactions.
• Findings reveal novel viral perturbations and dynamics associated with IBD.